[2106.07682] Revisiting Model Stitching to Compare Neural Representations

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status

[2106.07682] Revisiting Model Stitching to Compare Neural Representations Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Machine Learning arXiv:2106.07682 (cs) [Submitted on 14 Jun 2021] Title: Revisiting Model Stitching to Compare Neural Representations Authors: Yamini Bansal , Preetum Nakkiran , Boaz Barak View a PDF of the paper titled Revisiting Model Stitching to Compare Neural Representations, by Yamini Bansal and 2 other authors View PDF Abstract: We revisit and extend mode

related reading

• Model Stitchingarxiv.org
• Toy Models of Superpositiontransformer-circuits.pub
• The Platonic Representation Hypothesisphillipi.github.io
• Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
• Distill — Latest articles about machine learningdistill.pub
• The Platonic Representation Hypothesisarxiv.org
• The World Inside Neural Networksgoodfire.ai
• Formation of Representations in Neural Networksarxiv.org
• H-Nets - the Past | Goomba Labgoombalab.github.io
• An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
• Feature Visualizationdistill.pub
• Neural Networks, Types, and Functional Programming -- colah's blogcolah.github.io